Average Error: 22.9 → 22.9
Time: 1.3s
Precision: binary64
\[\frac{fyy \cdot fx - fxy \cdot fy}{fxx \cdot fyy - fxy \cdot fxy}\]
↓
\[\frac{fyy \cdot fx - fxy \cdot fy}{fxx \cdot fyy - fxy \cdot fxy}\]
\frac{fyy \cdot fx - fxy \cdot fy}{fxx \cdot fyy - fxy \cdot fxy}
↓
\frac{fyy \cdot fx - fxy \cdot fy}{fxx \cdot fyy - fxy \cdot fxy}
double code(double fyy, double fx, double fxy, double fy, double fxx) {
	return ((double) (((double) (((double) (fyy * fx)) - ((double) (fxy * fy)))) / ((double) (((double) (fxx * fyy)) - ((double) (fxy * fxy))))));
}
↓
double code(double fyy, double fx, double fxy, double fy, double fxx) {
	return ((double) (((double) (((double) (fyy * fx)) - ((double) (fxy * fy)))) / ((double) (((double) (fxx * fyy)) - ((double) (fxy * fxy))))));
}

Error

Bits error versus fyy

Bits error versus fx

Bits error versus fxy

Bits error versus fy

Bits error versus fxx

Try it out

Your Program's Arguments

Results

Enter valid numbers for all inputs

Derivation

  1. Initial program 22.9

    \[\frac{fyy \cdot fx - fxy \cdot fy}{fxx \cdot fyy - fxy \cdot fxy}\]

  2. Final simplification22.9

    \[\leadsto \frac{fyy \cdot fx - fxy \cdot fy}{fxx \cdot fyy - fxy \cdot fxy}\]

Reproduce

herbie shell --seed 2020152 
(FPCore (fyy fx fxy fy fxx)
  :name "(/ (- (* fyy fx) (* fxy fy)) (- (* fxx fyy) (* fxy fxy)))"
  :precision binary64
  (/ (- (* fyy fx) (* fxy fy)) (- (* fxx fyy) (* fxy fxy))))